Development of Digital Twin Model of Gas Turbine for Performance Simulation

EasyChair Preprint no. 13026

Abstract

In the context of escalating global demands for clean energy and the imperative to mitigate environmental pollution, heavy-duty gas turbines are assuming an increasingly pivotal role in the energy landscape. Renowned for their low emissions, high thermal efficiency, and adaptable start-up capabilities, these turbines are indispensable assets in the pursuit of sustainable power generation. Accurate modeling and simulation of gas turbine performance are paramount for the precise design of large-scale industrial turbines. Furthermore, timely monitoring of performance degradation is essential for fault diagnosis, optimal efficiency, and predictive maintenance. However, achieving this task is inherently challenging due to the complexities of turbine operation, including high nonlinearity, system intricacies, varying conditions, and the interplay of numerous parameters within the harsh operational environment. As ambient conditions fluctuate, major turbine components such as the compressor, combustor, and turbine may experience performance degradation over time, necessitating performance-based maintenance to maximize productivity. The efficacy of such maintenance practices hinges on real-time efficiency monitoring for timely diagnosis and prognosis of impending deterioration.To enhance the reliability and availability of gas turbines across diverse operating conditions, accurate and efficient simulation of turbine performance is indispensable. This forms the foundation for effective fault diagnosis and predictive maintenance strategies. In this paper, we introduce a novel approach that leverages the concept of a digital twin—a virtual replica of the physical turbine.e.

Keyphrases: Gas turbine, thermodynamic model, transient performance

@Booklet{EasyChair:13026,
  author = {Ashish Yadav and Crs Kumar},
  title = {Development of Digital Twin Model of Gas Turbine for Performance Simulation},
  howpublished = {EasyChair Preprint no. 13026},

  year = {EasyChair, 2024}}